Datasets:

Project-AgML
/

tomato_factory_detection

dataset_info:
  features:
    - name: image
      dtype: image
    - name: objects
      struct:
        - name: bbox
          list:
            list: float64
        - name: categories
          list:
            class_label:
              names:
                '0': green
                '1': red
  splits:
    - name: train
      num_bytes: 3522691102
      num_examples: 520
  download_size: 3522739307
  dataset_size: 3522691102
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
license: cc-by-4.0
task_categories:
  - object-detection
size_categories:
  - n<1K

Tomato Factory Detection

A dataset for detection of tomatoes in a plant factory setting. The dataset contains 520 images with 8,223 bounding box annotations across 2 categories.

This dataset is indexed on https://project-agml.github.io/ as part of the AgML python library.

Citation

@article{wu2023dataset,
  title={A dataset of tomato fruits images for object detection in the complex lighting environment of plant factories},
  author={Wu, Zhen-wei and Liu, Ming-hao and Sun, Cheng-xiu and Wang, Xin-fa},
  journal={Data in Brief},
  volume={48},
  pages={109291},
  year={2023},
  publisher={Elsevier}
}

Wu, Zhenwei; Wang, Xinfa; Liu, Minghao; Sun, Chengxiu (2026), “TomatoPlantfactoryDataset”, Mendeley Data, V3, doi: 10.17632/8h3s6jkyff.3